MS, University of Cincinnati, 2024, Engineering and Applied Science: Biomedical Engineering
The use of ultrasound in speech therapy provides real-time visual feedback of tongue movement during speech. Ultrasound is a therapeutic tool for treating sounds with complex articulatory patterns and diverse populations, including individuals with speech sound disorders, hearing impairments, childhood apraxia of speech, and speech errors following a stroke. The benefits of ultrasound biofeedback therapy (UBT) for remediating speech deficits could be enhanced by integrating UBT into a real-time, gamified interface that provides real-time articulatory feedback and promotes an external focus of attention, reducing the complex cognitive demands required for standard UBT. Previous studies have shown that accuracy of American English rhotic /r/ can be predicted by a single parameter, d, the difference between tongue dorsum and blade displacements measured by ultrasound imaging during speech production. This parameter has classified speech productions of rhotic syllables as correct versus misarticulated with a classification accuracy up to 85%. However, implementation of this classification approach into real-time gamified UBT, including both measurement timing and establishment of difficulty levels for progressive therapy, would benefit from optimization that is clinically relevant to real-time gamified UBT and uses a larger dataset than previously analyzed. 2,300 productions of 10 distinct rhotic syllables (including prevocalic and postvocalic contexts) from 49 children, with and without speech errors, were analyzed. For each production, ultrasound image sequences were processed by TonguePART software to acquire tongue displacement trajectories, and accuracy was judged by trained listeners using a visual analog scale (VAS). For each syllable, receiver operating characteristic (ROC) curve analyses using an 8-fold cross validation method were conducted to optimize selection of the image frame approach for classification using the d parameter and determine parameter thresh (open full item for complete abstract)
Committee: T. Douglas Mast Ph.D. (Committee Chair); Michael Riley Ph.D. (Committee Member); Kevin Haworth Ph.D. (Committee Member); Suzanne Boyce Ph.D. (Committee Member)
Subjects: Biomedical Research